Internet Protocols Fall Lectures 7-8 Network Layer Andreas Terzis

Transcription

1 Internet Protocols Fall 2006 Lectures 7-8 Network Layer Andreas Terzis

2 Outline Internet Protocol Service Model Addressing Original addressing scheme Subnetting CIDR Forwarding Router Designs Fragmentation ICMP Address Shortage NAT IPv6 CS 349/Fall06 2

3 IP Internet Concatenation of Networks Three requirements Connectivity Addressing Path discovery H1 H2 H3 Network 2 (Ethernet) R1 H4 Network 3 (FDDI) H5 H6 R2 Network 1 (Ethernet) H7 R3 H8 Network 4 (point-to-point) H1 H8 TCP R1 R2 R3 TCP IP IP IP IP IP ETH ETH FDDI FDDI PPP PPP ETH ETH CS 349/Fall06 3

4 Service Model Connectionless (datagram-based) Best-effort delivery (unreliable service) packets are lost packets are delivered out of order duplicate copies of a packet are delivered packets can be delayed for a long time Datagram format Version HLen TOS Length Ident Flags Offset TTL Protocol Checksum SourceAddr DestinationAddr Options (variable) Data Pad (variable) App Transport Network Link TCP / UDP IP Data Data Hdr TCP Segment Hdr IP Datagram CS 349/Fall06 4

5 Global Addresses Properties globally unique hierarchical: network + host Why hierarchy? Dot Notation (a) 0 Network Host (b) 1 0 Network Host 21 8 (c) Network Host CS 349/Fall06 5

6 Address Allocation Who manages the IP address space? Hierarchical system Internet Assigned Numbers Authority (IANA) Local Internet Registries (LIRs) ISPs ARIN, APNIC, etc CS 349/Fall06 6

7 Datagram Forwarding (v.1) Strategy every datagram contains destination s address if connected to destination network, then forward to host if not directly connected, then forward to some router forwarding table maps network number into next hop Network 1 (Ethernet) Example (R2) Network Number Next Hop 1 R3 H1 H2 Network 2 (Ethernet) H3 R1 H7 R3 H8 Network 4 (point-to-point) 2 R1 3 interface 1 H4 Network 3 (FDDI) R2 4 interface 0 H5 H6 CS 349/Fall06 7

8 IP Addressing Problem: Address classes were too rigid. Organizations with internal routers needed to have a separate (Class C) network ID for each link. And then every other router in the Internet had to know about every network ID in every organization, which led to large address tables. Small organizations wanted Class B in case they grew to more than 255 hosts. But there were only about 16,000 Class B network IDs. CS 349/Fall06 8

9 IP Addressing Two solutions were introduced: Subnettingis used within an organization to subdivide the organization s network ID. Classless Interdomain Routing (CIDR) CS 349/Fall06 9

10 Subnetting Add another level to address/routing hierarchy: subnet Subnet masks define variable partition of host part Subnets visible only within site Network number Host number Class B address Subnet mask ( ) Network number Subnet ID Host ID Subnetted address CS 349/Fall06 10

11 Subnet Example Subnet mask: Subnet number: H1 R Subnet mask: Subnet number: H Subnet mask: Subnet number: R H Forwarding table at router R1 Subnet Number Subnet Mask Next Hop interface interface R2 CS 349/Fall06 11

12 Forwarding Algorithm (v.2) D = destination IP address for each entry (SubnetNum, SubnetMask, NextHop) D1 = SubnetMask & D if D1 == SubnetNum if NextHop is an interface deliver datagram directly to D else deliver datagram to NextHop Use a default router if nothing matches Subnets not visible from the rest of the Internet CS 349/Fall06 12

13 Classless Interdomain Routing (CIDR) The IP address space is broken into line segments. Each line segment is described by a prefix. A prefix is of the form x/y where x indicates the prefix of all addresses in the line segment, and y indicates the length of the segment. e.g. The prefix 128.9/16 represents the line segment containing addresses in the range: / / / CS 349/Fall06 13

14 Classless Interdomain Routing (CIDR) / / / / / Most specific route = longest matching prefix CS 349/Fall06 14

15 Classless Interdomain Routing (CIDR) Prefix aggregation: If a service provider serves two organizations with prefixes, it can (sometimes) aggregate them to form a larger prefix. Other routers can refer to this larger prefix, and so reduce the size of their address table. E.g. ISP serves /24 and /24, it can tell other routers to send it all packets belonging to the prefix /23. ISP Choice: In principle, an organization can keep its prefix if it changes service providers. CS 349/Fall06 15

16 Hierarchical addressing: route aggregation Hierarchical addressing allows efficient advertisement of routing information: Organization /23 Organization /23 Organization /23 Organization Fly-By-Night-ISP Send me anything with addresses beginning /20 Internet /23 ISPs-R-Us Send me anything with addresses beginning /16 CS 349/Fall06 16

17 Hierarchical addressing: route aggregation Organization /23 Organization /23 Organization /23 Organization 7 Multi-homing / Fly-By-Night-ISP ISPs-R-Us Send me anything with addresses beginning /20 Send me anything with addresses beginning /16, /16 or /23 Internet CS 349/Fall06 17

18 How a Router Forwards Datagrams R R2 R3 R4 e.g => Port 2 Prefix Next-hop Port 65/ / / / / / / Forwarding table CS 349/Fall06 18

19 Forwarding in an IP Router (v.3) Lookup packet DA in forwarding table. If known, forward to correct port. If unknown, drop packet. Decrement TTL, update header Checksum. Forward packet to outgoing interface. Transmit packet onto link. Question: How is the address looked up in a real router? CS 349/Fall06 19

20 Lookup Performance Required Line Line Rate Pkt-size=40B Pkt-size=240B T1 1.5Mbps 4.68 Kpps 0.78 Kpps OC3 155Mbps 480 Kpps 80 Kpps OC12 622Mbps 1.94 Mpps 323 Kpps OC48 2.5Gbps 7.81 Mpps 1.3 Mpps OC Gbps Mpps 5.21 Mpps CS 349/Fall06 20

21 Data Address Direct Lookup IP Address Memory Next-hop, Port Problem: With 2 32 addresses, the memory would require 4 billion entries. CS 349/Fall06 21

22 Associative Lookups Contents addressable memory (CAM) Advantages: Associative Memory or CAM Simple Disadvantages Search Data 32 Network Address Port Number Port Number Hit? Slow High Power Small Expensive CS 349/Fall06 22

23 Hashed Lookups Search Data 32 Hashing Function 16 Address Memory Data Associated Data Hit? {Address log 2 N CS 349/Fall06 23

24 Lookups Using Hashing An example Memory #1 #2 #3 #4 Search Data 32 Hashing Function 16 #1 #2 Associated Data Hit? Linked list of entries with same hash key. #1 #2 #3 CS 349/Fall06 24

25 Lookups Using Hashing Advantages: Simple Expected lookup time can be small Disadvantages Non-deterministic lookup time Inefficient use of memory CS 349/Fall06 25

26 Patricia Tries Example Prefixes: d e 0 1 f g h i Skip 5 j a) b) c) d) 001 e) 0101 f) 011 g) 100 h) 1010 i) 1100 j) a bc CS 349/Fall06 26

27 How to Implement a router Issues Performance Throughput Scaling CS 349/Fall06 27

28 Workstation-Based Aggregate bandwidth 1/2 of the I/O bus bandwidth capacity shared among all hosts connected to router example: 1Gbps bus can support 5 x 100Mbps ports (in theory) Packets-persecond must be able to switch small packets 300,000 packets-persecond is achievable e.g., 64-byte packets implies 155Mbps CPU Main memory I/O bus Interface 1 Interface 2 Interface 3 CS 349/Fall06 28

29 Switching Hardware Design Goals throughput (depends on traffic model) scalability (a function of n) Control processor Ports buffering (input and/or output) Fabric as simple as possible sometimes do buffering (internal) Input port Switch fabric Output port CS 349/Fall06 29

30 Fragmentation Problem: A router may receive a packet larger than the maximum transmission unit (MTU) of the outgoing link. Source Destination Ethernet A MTU=1500 bytes MTU=1500 bytes B R1 MTU<1500 bytes R2 Solution: R1 fragments the IP datagram into mutiple, self-contained datagrams. Offset>0 More Frag=0 Data HDR (ID=x) Offset=0 More Frag=1 Data HDR (ID=x) Data HDR (ID=x) Data HDR (ID=x) CS 349/Fall06 30

31 Fragmentation (II) Fragments are re-assembled by the destination host; not by intermediate routers. To avoid fragmentation, hosts commonly use path MTU discovery to find the smallest MTU along the path. Path MTU discovery involves sending various size datagrams until they do not require fragmentation along the path. Most links use MTU>=1500bytes today. Try: traceroute f and traceroute f (DF=1 set in IP header; routers send ICMP error message, which is shown as!f ). CS 349/Fall06 31

32 ICMP Internet Control Message Protocol: Used by a router/end-host to report some types of error: E.g. Destination Unreachable: packet can t be forwarded to/towards its destination. E.g. Time Exceeded: TTL reached zero, or fragment didn t arrive in time. Traceroute uses this error to its advantage. An ICMP message is an IP datagram, and is sent back to the source of the packet that caused the error. CS 349/Fall06 32

33 IP Address Shortage Global IPv4 addresses are getting depleted Increase in number of hosts PCs, PDAs, cellphones, microwaves, etc Address inefficiencies What to do Get larger address space -> IPv6 Remove the assumption that address is globally unique Can reuse the same address multiple times -> NAT CS 349/Fall06 33

34 NAT: Network Address Translation rest of Internet local network (e.g., home network) / All datagrams leaving local network have same single source NAT IP address: , different source port numbers Datagrams with source or destination in this network have /24 address for source, destination (as usual) CS 349/Fall06 34

35 NAT: Network Address Translation 2: NAT router changes datagram source addr from , 3345 to , 5001, updates table 2 NAT translation table WAN side addr LAN side addr , , 3345 S: , 5001 D: , S: , 3345 D: , : host sends datagram to , S: , 80 D: , : Reply arrives dest. address: , 5001 S: , 80 D: , : NAT router changes datagram dest addr from , 5001 to , 3345 CS 349/Fall06 35

36 NAT implementation NAT router must: outgoing datagrams: replace (source IP address, port #) of every outgoing datagram to (NAT IP address, new port #)... remote clients/servers will respond using (NAT IP address, new port #) as destination addr. remember (in NAT translation table) every (source IP address, port #) to (NAT IP address, new port #) translation pair incoming datagrams: replace (NAT IP address, new port #) in dest fields of every incoming datagram with corresponding (source IP address, port #) stored in NAT table CS 349/Fall06 36

37 NAT Problems Problems due to NAT Increased network complexity, reduced robustness Cannot run services inside NAT (maybe) Address shortage should instead be solved by IPv6 CS 349/Fall06 37

38 IPv6 Motivation: 32-bit address space exhaustion Take the opportunity for some clean-up IPv6 datagram format: fixed-length 40 byte header Address length changed from 32 bits to 128 bits fragmentation fields moved out of base header IP options moved out of base header Header Length field eliminated Header Checksum eliminated Type of Service field eliminated Time to Live Hop Limit, Protocol Next Header Precedence Priority, added Flow Label field Length field excludes IPv6 header CS 349/Fall06 38

39 IPv6 header format Version Priority Flow Label Payload Length Next Header Hop Limit Source Address (16 bytes, 128 bits) Destination Address (16 bytes) IPv4 header Version Hdr Len Prec TOS Total Length Identification Flags Fragment Offset Time to Live Protocol Header Checksum Source Address Destination Address Options Padding 32 bits CS 349/Fall06 39

40 Transition From IPv4 To IPv6 Not all routers can be upgraded simultaneously Two proposed approaches to allow the Internet operate with mixed IPv4 and IPv6 routers : Tunneling: IPv6 carried as payload in IPv4 packets among IPv4 routers Dual Stack: some routers with dual stack (v6, v4) can translate between formats CS 349/Fall06 40